My friend Evgeny Morozov is one of the most insightful technology journalists working today, writing for The Economist, BusinessWeek and Le Monde. (That’s what I think even on weeks where he hasn’t written extremely kind things about my projects.) His blog, dominated by long lists of consistently interesting bookmarks, is one of my few daily reads, in part because he’s one of my main sources of serendipity, finding things interesting to me, often on topics I didn’t know I was interested in.
Evgeny is deeply concerned with the question of serendipity and on the phenomenon of media cocooning, the tendency of people to surround themselves with media that echoes topics, interests and political points of view that we share. More to the point, as a Belarussian fascinated both international politics, he’s frustrated that popular blog aggregators rarely track news from around the world, or news on subjects other than technology. From an email he sent to friends this morning:
Over the years of following the English-language blogosphere, I have become increasingly frustrated with the absence of news aggregators that could help me stay on top of important developments in non-tech areas. Fields like economics, design, law, environment, or literature didn’t seem to have their own Digg, Techmeme or Technorati; thus, navigating through the growing non-tech blogospheres has become very difficult. As the amount of information on the Web has kept growing rapidly, it has proved quite challenging to remain a true polymath, i.e. remain continuously well-informed about a multitude of fields, not just one.
Fighting fire with fire, Evgeny’s released a new aggregator, Polymeme. Using software he and a team of developers have put together, the system tracks 20 collections of blogs, each tightly focused on a specific topic, like Economics or Evolution. The software reads the RSS feeds of the blogs and discovers what news stories, blogposts or other media that community is pointing to and discussing. The system offers stories and blogposts clustered around the topics, which are paired with (creative-commons licensed, discovered on Flickr) photos and organized on a frontpage and subject pages. Other tools allow you to find popular memes, phrases and topics that have been identified across communities.
Most of the stories Polymeme finds, Evgeny tells me, are mainstream media stories… just ones that you’re not likely to find via Digg and Reddit – instead, they’re the stories getting discussed by economists, evolutionary biologists or other smart, subject-focused bloggers. It’s my experience that bloggers focused on a particular issue – Afrophilia, for instance – tend to flock to a set of articles that become important discussion points in that sphere, even if they’re invisible to the rest of the web as a whole. Polymeme gives you the chance to listen in on some of those conversations, even if you’re not an active blogger in that space.
Some things I think are great about the idea and the early incarnation of the system:
- Using bloggers as a set of experts to find relavent content is very, very smart, as that’s what bloggers are already do, and because human filtering systems are much more powerful than purely algorithmic ones.
- This is one of the first tools I’ve seen with an explicit promise to diversify voices on the web and break out of existing echo chambers. From a description on the site of the project: “Polymeme helps you discover intelligent content that lies beyond the usual echo chambers of tech news, celebrity gossip or American politics.”
- Because the collection of blogs the system uses is quite diverse, the sources cited tend to be significantly broader than those I see on other aggregators.
- The site is organized in terms of a central story, followed by stories and blogposts that comment on it, making it quite easy to find a path into a conversation on a breaking topic.
Spme open questions I’m curious to explore:
- Will the topics the system covers be as insular and echo-filled as the tech and US politics blogosphere? Is there a danger that Polymeme is just making more echo chambers?
- Is clustering stories enough, or will Polymeme need to do some storytelling to help encourage people to explore these new stories?
- Will the system work without the sorts of community voting functions that Digg and Reddit rely on? Are bloggers a better quality filter than a reader community?
I’m thrilled that Evgeny is trying a practical response to challenges about homophily and serendipity and fascinated to see where this will go. I hope you’ll give it a try.